PROJECT SUMMARY/ABSTRACT Extensively drug-resistant tuberculosis (XDR-TB), which has been associated with up to 80% mortality, is considered ?virtually untreatable? in many parts of the world, and is threatening global TB control. Patients with any form of DR-TB have a much greater risk of dying than those with drug susceptible TB; resistance to second-line anti-TB drugs increases this risk more than fourfold, and treatment success rates can be as low as 30%. We have developed an optimized Targeted Next Generation Sequencing (NextGen-RDST) DST which can be performed directly from sputum samples. Pilot data of Version 1 of this test indicates it has excellent performance to detect MDR-TB (sensitivity of 97.6%, specificity of 98.9%) and XDR-TB (sensitivity of 90.0%, specificity of 97.8%). The goal of this project is to leverage well-characterized clinical sputum samples that were prospectively collected from an existing NIH project examining patients at risk for MDR-TB (R01AI111435, PI Rodwell) to verify the performance of our optimized NextGen-RDST. There is currently no FDA approved rapid diagnostic for XDR-TB, which is why the 2013 CDC fact sheet states that diagnosis of XDR-TB (a NIAID Category C priority pathogen) currently requires ?6 to 16 weeks? to complete. There is a clear and critical need for an integrated, table-top platform for high sensitivity, high specificity, rapid diagnosis of drug resistant TB directly from patient samples. We propose to combine the experience, resources and existing diagnostic testing pipeline of the Global Consortium for Drug Resistant Tuberculosis Diagnostics (NIAID U01AI082229 and NIAID R01AI111435) with our project partner, Translational Genomics Research Institute (TGen) to further develop our optimized NextGen-RDST as a rapid XDR-TB diagnostic platform. Based on our extensive experience with the genetic basis of XDR-TB, and TGen's prototype platform for identifying specific single nucleotide polymorphisms (SNPs) directly from clinical samples, we hypothesize that we will be able to detect XDR-TB isolates with clinically relevant phenotypic resistance to isoniazid, rifampin, fluoroquinolones and injectable anti-TB drugs with 90-98% sensitivity and ~100% specificity, and that discordances between a phenotypic standard and NextGen-RDST can be appropriately resolved through further genomic experimentation. We will achieve this goal by completing the following three objectives; (1) determining the sensitivity and specificity of our novel direct NextGen-RDST relative to reference standard phenotypic DST from cultures of the same sputa, collected from 500 patients at risk for MDR-TB in Moldova; (2) establishing the role that a culture step plays in the discordance between reference standards and novel NextGen-RDST; (3) determining the role subpopulations have in the discordance between the reference standards and the novel NextGen-RDST.